Ethernet OAM (Operation Administration and Maintenance) used for operation administration and maintenance in Ethernet™ is specified by ITU-T (International Telecommunication Union Telecommunication Standardization Sector) Y.1731 and IEEE (Institute of Electrical and Electronic Engineers) 802.1ag. In Ethernet OAM, a supervision section is called an ME (Maintenance Entity), and an ME can be set for every MEG (Maintenance Entity Group) level. As MEG levels, eight supervision levels 0 to 7 are specified. Moreover, there is a need to set MEPs (Maintenance entity group End Points) as end points at both ends of a supervision section (an ME).
In FIG. 15, a setting example of MEs and MEPs is shown. As shown in FIG. 15, an ME (a supervision section) is set between devices, and MEPs (end points) shown by black inversed triangles are set at both ends of the supervision section.
Further, an ME and an MEP have an MEG level as an attribute, and are specified so that an ME with a lower MEG level does not overlie an ME with a higher MEG level as shown in FIG. 15.
Therefore, in the case of operating an Ethernet OAM frame to perform operation administration and maintenance, there is a need to preliminarily register, into each of Ethernet communication devices at MEP positions of each supervision section (ME), an MEP in an MEG level of the ME. Moreover, there is a case that an MIP (Maintenance Intermediate Point) that relays a supervision frame is registered in an Ethernet communication device connecting between MEPs.
Further, for shortening a fault detection time from occurrence of a unidirectional link fault, there is a device that transmits and receives link normality check frames that are specifically defined therefor and, when a destination table is searched based on destination information of the frames and consequently a corresponding entry does not exist, newly acids a destination to the destination table as a general learning process (e.g., refer to Patent Document 1).
Further, there is a device that, regarding a link aggregation management table managing the configuration of link aggregation in a layer-2 switch, when a fault occurs in a port belonging to one link aggregation group, causes another line card having the port belonging to the one link aggregation group to update the link aggregation management table (e.g., refer to Patent Document 2).    [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2009-135625    [Patent Document 2] Japanese Unexamined Patent Application Publication No. 2008-244907
However, since there is a need to individually set registration information into the respective Ethernet communication devices in the MEP positions of each ME in a network at the time of the aforementioned MEP registration, it is required to set by connecting a maintenance terminal to each of the Ethernet communication devices or connect a memory card containing registration information to each of the Ethernet communication devices and load the registration information.
Consequently, for registration of an MEP, each of the Ethernet communication devices needs to be equipped with an interface and CPU (Central Processing Unit) for setting of registration information, and there is a fear that a circuit becomes large-scale and power consumption and cost increase.
Further, since the registration information of the MEP to be registered into each of the Ethernet communication devices is in large amounts and complicated, there is also a fear of wrong setting of the registration information.
Further, addition of a destination to the destination table in the aforementioned Patent Document 1 is simple increase of the number of destinations in the table to shorten a fault detection time and is performed as a general learning process, and setting of a device is not performed in consideration of simplification of a circuit or reduction of cost.
Further, in order to shorten a fault detection time, specifically defined link normality check frames are transmitted and received, but periodic use of supervision frames is not taken into consideration.
Further, the aforementioned Patent Document 2 shows a table update operation when a fault is detected in transmission and reception of usual frames, but use of supervision frames periodically transmitted and received or automatic setting of a device is not take into consideration.